Hello Sympy Developers,
I have recently started planning the design and integration of the *Power Series domain* and would appreciate some guidance on key structural decisions. Currently, my implementation revolves around three primary classes: - *Domain* - *Ring* - *Element* However, as I proceed further, I have a few questions: *1. Where should core functionality be implemented?* - Should functions like *exp, sin, cos* be implemented natively in the *Element* class, or should we rely on the existing *ring_series* implementation? - If we use *ring_series*, what would be the best way to handle cases where a function is not well-defined due to the presence of a *constant term* in the expression? *2. Additional Questions* - Currently, multivariate expansion considers one term as a variable while treating the rest as constants. Would implementing a more generalized multivariate expansion be feasible within the project timeline? If so, how should constant terms be handled in such an expansion? - Would introducing *more trigonometric functions* in *ring_series* be a good idea? I would appreciate any insights on these design choices to ensure better integration with SymPy’s existing structure. Best regards, *Jatin Bhardwaj* On Sunday, 16 February 2025 at 17:30:01 UTC+5:30 Jatin Bhardwaj wrote: > Hello SymPy developers, > > I am interested in contributing to SymPy by implementing a power series > domain, specifically formal power series rings (e.g., QQ[[x]]). > Currently, the ring_series module provides low-level operations, but > there is no dedicated domain representation for power series. Integrating > this with SymPy’s domain system—similar to how python-flint handles > fmpq_series—could significantly enhance symbolic series computations. > > There is an ongoing discussion on this topic in gh-26957 > <https://github.com/sympy/sympy/issues/26957>, which provides a strong > starting point for this project. The discussion outlines key considerations > and potential approaches, making it a well-structured project to work on. > > I had previously written a proposal for improving PDE-solving > functionality in SymPy. Which of these two would better fit SymPy’s roadmap > and priorities? I would also appreciate any further guidance on design > choices, integration challenges, or other considerations not covered in the > discussion. > > > Thank you for your time and consideration. I look forward to your response. > > Jatin Bhardwaj > -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/sympy/1cfec4b9-1826-44b6-824f-f15f7f7d3470n%40googlegroups.com.
